I'm currently have a longitudinal dataset with dependent variable Y measures over two time points (variable = Time) across 4 groups (variable = Group). This variable Y is a continuous variable describing brain structural properties. The range for this variable can differ based on what property is being analyzed (from between around 2000-9000 or from between around 1-10). I was planning to run separate models for each variable Y.

I want to take account the variability of each subjects (variable = Subject) intercept when looking at a linear fit across the two time points. I want to do this across all the models with different Y variables.

I am looking to use a mixed model approach to determine the prediction of Y while controlling for two continuous covariates, which are age and another continuous structural brain property which is usually in the 5 digit range (variables B + C).

Currently I am using the lme4 package and the lmer() function as below

lmer(Y ~ Group*Time + B + C + (1 + Time | Subject), data = data)

Is the correct model to accurately model the differences in groups over time while taking to account subject variability?

Thanks in advance!

  • $\begingroup$ Do I understand correctly that you are more interested in individual level predictions and not so much understanding group means? $\endgroup$ – StatsStudent May 20 at 19:56
  • $\begingroup$ I am interested in the group level means, but do want to take account individual level variability as well. Thanks! $\endgroup$ – Ravi May 20 at 20:21
  • $\begingroup$ Can you describe your dependent variable a bit more? Is it continuous or binary? Perhaps ordinal? $\endgroup$ – StatsStudent May 20 at 20:42
  • $\begingroup$ The dependent variable is a continuous variable. $\endgroup$ – Ravi May 20 at 20:44
  • $\begingroup$ I was really hoping for some additional insight on the variable. What is it describing? What's it's range? Can you edit your initial question and provide significantly more detail? $\endgroup$ – StatsStudent May 21 at 1:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.